knowing where events are going to happen

20
©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential Knowing where events are going to happen CANE – September 2014 Howard A Kunst, FCAS MAAA Chief Actuary, CoreLogic Spatial Solutions

Upload: tejana

Post on 04-Jan-2016

36 views

Category:

Documents


3 download

DESCRIPTION

Knowing where events are going to happen. Howard A Kunst, FCAS MAAA Chief Actuary, CoreLogic Spatial Solutions. CANE – September 2014. Antitrust Notice. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Knowing where events are going to happen

CANE – September 2014

Howard A Kunst, FCAS MAAA

Chief Actuary, CoreLogic Spatial Solutions

Page 2: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

The Casualty Actuarial Society is committed to adhering strictly to the letter and spirit of the antitrust laws. Seminars conducted under the auspices of the CAS are designed solely to provide a forum for the expression of various points of view on topics described in the programs or agendas for such meetings.

Under no circumstances shall CAS seminars be used as a means for competing companies or firms to reach any understanding –expressed or implied – that restricts competition or in any way impairs the ability of members to exercise independent business judgment regarding matters affecting competition.

It is the responsibility of all seminar participants to be aware of antitrust regulations, to prevent any written or verbal discussions that appear to violate these laws, and to adhere in every respect to the CAS antitrust compliance policy.

Antitrust Notice

Page 3: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Session Objectives

What determines the short term vs long term frequency? Geographical accuracy

Models predict where the events are more likely to occur Geocoding accuracy – understanding granular data

Validation of recent history of events Hail Tornado Wildfire

Combination of multiple perils – Single Risk Score

3

Page 4: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Understanding where events can happen

What affects the short term frequency of events? Climate change – frequency of events and geographic distribution Solar activity Heat – sunk in the earth, stored in oceans Drought conditions / Abundance of precipitation Time elapsed since last event Random variation

It’s almost impossible to predict exactly when and where the next event is going to happen But if you can predict the locations with the highest risk for an event, isn’t

that just as much or more important for setting a natural catastrophe plan/strategy?

4

Page 5: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Knowing where events are going to happen

Scientists gather information on the both short/long term factors that increase/decrease the likelihood of natural catastrophic events

Models are developed that set a risk level for every location across the US – extremely granular (10 meter grids)

Revisions are made annually to adjust for more recent information (ie vegetation changes / drought conditions for Wildfires)

Ongoing validation on the accuracy of the models by comparing where recent events have occurred to the estimated risk levels Do the actual events in fact occur in the higher risk areas that we

expected?

5

Page 6: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

The importance of granularity

Parcel boundary data represents the legal extents of each taxable U.S. property address.

There are an estimated 145 million privately owned parcels in the U.S.

CoreLogic has converted and normalized about 140 million parcels from state, county, city, and town sources

As these digital parcel boundaries become available they are rapidly being incorporated into applications to enhance: o Geocoding accuracyo Risk assessmento Risk concentrationo Many other uses where “granular” accuracy is

important

What is Parcel Data?

Page 7: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Wildfire Risk Determination

Data Elements

Digital Elevation Model (DEM) Satellite Imagery Vegetation Condition Class data

Data Granularity

Input cell size based on 30m grid All layers sampled at 30m

Vegetation

Composition Class

Aspect

Slope

Wildfire Risk

20% of Total Risk

80% of Total Risk

Slope

Aspect

Fuel Load

Condition Class

Topogra

phy

Fuel

Wild

fire

Risk

Page 8: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

NC QL2 LiDAR (2014)

Actual Survey Elevations

30m Profile

10m Profile4m Profile

QL2 – 0.3m Profile

QL2 nearly mirrors existing high precision survey data

Cross-Section

River Floodplain

Floodplain

Page 9: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Severe Convective Storms (SCS)

Severe Convective Storms (SCS) refer to Tornado, Hail and Straight Line Wind events

Prior to recent events, less attention given to SCS vs Hurricane and Earthquake Tornados and the storms that generate tornados account for more than

half of the insured catastrophic losses (57%) that occur each year in the U.S. – and extend beyond “Tornado Alley”

Frequency of observed events has increased Growing population Better observational tools (Doppler radar, etc.) Rising global temperature?

9

Page 10: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential 10

Page 11: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Distribution of 2”+ Hail events

11

52.4% of Damaging Hail events occur in areas with Extreme Hail Risk levels

85.6% of Damaging Hail events occur in areas of High, Very High or Extreme Hail Risk; those areas are only 40.7% of the land area of the USA.

Page 12: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential 12

Page 13: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Annual EF0-EF5 Tornadoes

13

Page 14: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Distribution of F2 and above Tornado events

30.3% of F2 and higher events occur in areas with Extreme Tornado Risk levels

84.3% of F2 and above events occur in areas of High, Very High or Extreme Risk; those areas are only 40.0% of the land area of the USA.

14

Page 15: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Overview - CoreLogic Wildfire Hazard Risk Report Drought conditions (only 5 states in the western U.S. recorded

precipitation at or above average in 2012) and high temperatures impact the number of wildfires

67,664 Wildfires in 2012 9,326,000 acres burned in 2012 For 2000-2008, on average over 2,500 structures destroyed each

year, compared with less than 1,000 per year for all prior decades Over 1,260,000 residential properties in the western U.S are

currently located in High or Very High Wildfire risk areas, with a value of more than $189 billion

15

Page 16: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Wildfire – Bastrop County, TX

16

Page 17: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Wildfire – Bastrop County, TX

17

Wildfire – Bastrop County, TX

Page 18: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Wildfire – Bastrop County TX

18

% of properties damaged increases with Wildfire Risk Score Damage % varies across Very High Risk scores, but increasing

Wildfire Risk Score summary

Total % in groupWF Score properties w/ damageScores 0-50 3,328 0.0%scores 51-60 1,089 0.6%scores 61-70 2,151 1.5%scores 71-80 3,180 7.2%scores 81-90 5,302 15.8%scores 91-100 1,781 18.7%Total 16,831 8.5%

% of properties damaged or destroyed, by WF score

0.00%0.60%

1.50%

7.20%

15.80%

18.70%

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

16.00%

18.00%

20.00%

0-50 51-60 61-70 71-80 81-90 91-100

Page 19: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

Hazard Risk Score Summary

19

Page 20: Knowing where events are going to happen

©2013 CoreLogic, Inc. All rights reserved. Proprietary & Confidential

There is an increasing number of events over the past decade

Location of recent events seem to follow where the models predict they will occur

Hail and Tornado damage extends and causes damage beyond just the traditional Tornado Alley

20

Summary